Balancing Value Underestimation and Overestimationwith Realistic Actor-Critic.
Sicen LiGang WangQinyun TangLiquan WangPublished in: CoRR (2021)
Keyphrases
- actor critic
- reinforcement learning
- policy gradient
- temporal difference
- approximate dynamic programming
- optimal control
- neuro fuzzy
- gradient method
- reinforcement learning algorithms
- function approximation
- policy iteration
- average reward
- model free
- reinforcement learning methods
- neural network
- evaluation function
- markov chain
- fuzzy logic
- learning algorithm